From adf58ed6870302fd87a8a0fa25f24c0e4e0716e9 Mon Sep 17 00:00:00 2001 From: Nicholas Nethercote Date: Thu, 31 Jul 2014 23:46:37 -0700 Subject: [PATCH] Represent cid chars using integers, not strings. cid chars are 16-bit unsigned integers. Currently we convert them to single-char strings when inserting them into the CMap, and then convert them back to integers when extracting them from the CMap. This patch changes CMap so that cid chars stay in integer format throughout, saving both time and space. When loading the PDF from issue #4580, this change reduces peak RSS from ~600 to ~370 MiB. It also improves overall speed on that PDF by ~26%, going from 724 ms to 533 ms. --- src/core/cmap.js | 54 +++++++++++++++++++++++++----------------- src/core/fonts.js | 15 ++++++------ test/unit/cmap_spec.js | 6 ++--- 3 files changed, 42 insertions(+), 33 deletions(-) diff --git a/src/core/cmap.js b/src/core/cmap.js index 446b11e42056b..4ffeb7700593c 100644 --- a/src/core/cmap.js +++ b/src/core/cmap.js @@ -199,6 +199,10 @@ var CMap = (function CMapClosure() { // where nBytePairs are ranges e.g. [low1, high1, low2, high2, ...] this.codespaceRanges = [[], [], [], []]; this.numCodespaceRanges = 0; + // Map entries have one of two forms. + // - cid chars are 16-bit unsigned integers, stored as integers. + // - bf chars are variable-length byte sequences, stored as strings, with + // one byte per character. this._map = []; this.vertical = false; this.useCMap = null; @@ -210,18 +214,23 @@ var CMap = (function CMapClosure() { this.numCodespaceRanges++; }, - mapRange: function(low, high, dstLow) { + mapCidRange: function(low, high, dstLow) { + while (low <= high) { + this._map[low++] = dstLow++; + } + }, + + mapBfRange: function(low, high, dstLow) { var lastByte = dstLow.length - 1; while (low <= high) { - this._map[low] = dstLow; + this._map[low++] = dstLow; // Only the last byte has to be incremented. dstLow = dstLow.substr(0, lastByte) + String.fromCharCode(dstLow.charCodeAt(lastByte) + 1); - ++low; } }, - mapRangeToArray: function(low, high, array) { + mapBfRangeToArray: function(low, high, array) { var i = 0, ii = array.length; while (low <= high && i < ii) { this._map[low] = array[i++]; @@ -229,6 +238,7 @@ var CMap = (function CMapClosure() { } }, + // This is used for both bf and cid chars. mapOne: function(src, dst) { this._map[src] = dst; }, @@ -302,7 +312,7 @@ var IdentityCMap = (function IdentityCMapClosure() { CMap.call(this); this.vertical = vertical; this.addCodespaceRange(n, 0, 0xffff); - this.mapRange(0, 0xffff, '\u0000'); + this.mapCidRange(0, 0xffff, 0); } Util.inherit(IdentityCMap, CMap, {}); @@ -522,7 +532,7 @@ var BinaryCMapReader = (function BinaryCMapReaderClosure() { case 2: // cidchar stream.readHex(char, dataSize); code = stream.readNumber(); - cMap.mapOne(hexToInt(char, dataSize), String.fromCharCode(code)); + cMap.mapOne(hexToInt(char, dataSize), code); for (i = 1; i < subitemsCount; i++) { incHex(char, dataSize); if (!sequence) { @@ -530,7 +540,7 @@ var BinaryCMapReader = (function BinaryCMapReaderClosure() { addHex(char, tmp, dataSize); } code = stream.readSigned() + (code + 1); - cMap.mapOne(hexToInt(char, dataSize), String.fromCharCode(code)); + cMap.mapOne(hexToInt(char, dataSize), code); } break; case 3: // cidrange @@ -538,8 +548,8 @@ var BinaryCMapReader = (function BinaryCMapReaderClosure() { stream.readHexNumber(end, dataSize); addHex(end, start, dataSize); code = stream.readNumber(); - cMap.mapRange(hexToInt(start, dataSize), hexToInt(end, dataSize), - String.fromCharCode(code)); + cMap.mapCidRange(hexToInt(start, dataSize), hexToInt(end, dataSize), + code); for (i = 1; i < subitemsCount; i++) { incHex(end, dataSize); if (!sequence) { @@ -551,8 +561,8 @@ var BinaryCMapReader = (function BinaryCMapReaderClosure() { stream.readHexNumber(end, dataSize); addHex(end, start, dataSize); code = stream.readNumber(); - cMap.mapRange(hexToInt(start, dataSize), hexToInt(end, dataSize), - String.fromCharCode(code)); + cMap.mapCidRange(hexToInt(start, dataSize), hexToInt(end, dataSize), + code); } break; case 4: // bfchar @@ -578,9 +588,9 @@ var BinaryCMapReader = (function BinaryCMapReaderClosure() { stream.readHexNumber(end, ucs2DataSize); addHex(end, start, ucs2DataSize); stream.readHex(charCode, dataSize); - cMap.mapRange(hexToInt(start, ucs2DataSize), - hexToInt(end, ucs2DataSize), - hexToStr(charCode, dataSize)); + cMap.mapBfRange(hexToInt(start, ucs2DataSize), + hexToInt(end, ucs2DataSize), + hexToStr(charCode, dataSize)); for (i = 1; i < subitemsCount; i++) { incHex(end, ucs2DataSize); if (!sequence) { @@ -592,9 +602,9 @@ var BinaryCMapReader = (function BinaryCMapReaderClosure() { stream.readHexNumber(end, ucs2DataSize); addHex(end, start, ucs2DataSize); stream.readHex(charCode, dataSize); - cMap.mapRange(hexToInt(start, ucs2DataSize), - hexToInt(end, ucs2DataSize), - hexToStr(charCode, dataSize)); + cMap.mapBfRange(hexToInt(start, ucs2DataSize), + hexToInt(end, ucs2DataSize), + hexToStr(charCode, dataSize)); } break; default: @@ -675,7 +685,7 @@ var CMapFactory = (function CMapFactoryClosure() { obj = lexer.getObj(); if (isInt(obj) || isString(obj)) { var dstLow = isInt(obj) ? String.fromCharCode(obj) : obj; - cMap.mapRange(low, high, dstLow); + cMap.mapBfRange(low, high, dstLow); } else if (isCmd(obj, '[')) { obj = lexer.getObj(); var array = []; @@ -683,7 +693,7 @@ var CMapFactory = (function CMapFactoryClosure() { array.push(obj); obj = lexer.getObj(); } - cMap.mapRangeToArray(low, high, array); + cMap.mapBfRangeToArray(low, high, array); } else { break; } @@ -704,7 +714,7 @@ var CMapFactory = (function CMapFactoryClosure() { var src = strToInt(obj); obj = lexer.getObj(); expectInt(obj); - var dst = String.fromCharCode(obj); + var dst = obj; cMap.mapOne(src, dst); } } @@ -725,8 +735,8 @@ var CMapFactory = (function CMapFactoryClosure() { var high = strToInt(obj); obj = lexer.getObj(); expectInt(obj); - var dstLow = String.fromCharCode(obj); - cMap.mapRange(low, high, dstLow); + var dstLow = obj; + cMap.mapCidRange(low, high, dstLow); } } diff --git a/src/core/fonts.js b/src/core/fonts.js index 383465b306a51..de69f5ca27f6f 100644 --- a/src/core/fonts.js +++ b/src/core/fonts.js @@ -3899,8 +3899,7 @@ var Font = (function FontClosure() { var cidToGidMap = properties.cidToGidMap || []; var cidToGidMapLength = cidToGidMap.length; properties.cMap.forEach(function(charCode, cid) { - assert(cid.length === 1, 'Max size of CID is 65,535'); - cid = cid.charCodeAt(0); + assert(cid <= 0xffff, 'Max size of CID is 65,535'); var glyphId = -1; if (cidToGidMapLength === 0) { glyphId = charCode; @@ -4370,10 +4369,10 @@ var Font = (function FontClosure() { var cMap = properties.cMap; toUnicode = []; cMap.forEach(function(charcode, cid) { - assert(cid.length === 1, 'Max size of CID is 65,535'); + assert(cid <= 0xffff, 'Max size of CID is 65,535'); // e) Map the CID obtained in step (a) according to the CMap obtained // in step (d), producing a Unicode value. - var ucs2 = ucs2CMap.lookup(cid.charCodeAt(0)); + var ucs2 = ucs2CMap.lookup(cid); if (ucs2) { toUnicode[charcode] = String.fromCharCode((ucs2.charCodeAt(0) << 8) + @@ -4415,7 +4414,7 @@ var Font = (function FontClosure() { var charcode = 0; if (this.composite) { if (this.cMap.contains(glyphUnicode)) { - charcode = this.cMap.lookup(glyphUnicode).charCodeAt(0); + charcode = this.cMap.lookup(glyphUnicode); } } // ... via toUnicode map @@ -4444,7 +4443,7 @@ var Font = (function FontClosure() { var widthCode = charcode; if (this.cMap && this.cMap.contains(charcode)) { - widthCode = this.cMap.lookup(charcode).charCodeAt(0); + widthCode = this.cMap.lookup(charcode); } width = this.widths[widthCode]; width = isNum(width) ? width : this.defaultWidth; @@ -5626,8 +5625,8 @@ var CFFFont = (function CFFFontClosure() { // If the font is actually a CID font then we should use the charset // to map CIDs to GIDs. for (glyphId = 0; glyphId < charsets.length; glyphId++) { - var cidString = String.fromCharCode(charsets[glyphId]); - var charCode = properties.cMap.charCodeOf(cidString); + var cid = charsets[glyphId]; + var charCode = properties.cMap.charCodeOf(cid); charCodeToGlyphId[charCode] = glyphId; } } else { diff --git a/test/unit/cmap_spec.js b/test/unit/cmap_spec.js index d690adb4012c5..2a4fb9cd57f23 100644 --- a/test/unit/cmap_spec.js +++ b/test/unit/cmap_spec.js @@ -44,7 +44,7 @@ describe('cmap', function() { 'endcidchar\n'; var stream = new StringStream(str); var cmap = CMapFactory.create(stream); - expect(cmap.lookup(0x14)).toEqual(String.fromCharCode(0x00)); + expect(cmap.lookup(0x14)).toEqual(0x00); expect(cmap.lookup(0x15)).toBeUndefined(); }); it('parses begincidrange', function() { @@ -54,8 +54,8 @@ describe('cmap', function() { var stream = new StringStream(str); var cmap = CMapFactory.create(stream); expect(cmap.lookup(0x15)).toBeUndefined(); - expect(cmap.lookup(0x16)).toEqual(String.fromCharCode(0x00)); - expect(cmap.lookup(0x1B)).toEqual(String.fromCharCode(0x05)); + expect(cmap.lookup(0x16)).toEqual(0x00); + expect(cmap.lookup(0x1B)).toEqual(0x05); expect(cmap.lookup(0x1C)).toBeUndefined(); }); it('decodes codespace ranges', function() {